Commit Graph

105 Commits

Author SHA1 Message Date
Xin LI
61613f1912 Utilize calloc() instead of bzero'ing manually. 2009-02-21 03:43:20 +00:00
Giorgos Keramidas
e9d8f84b73 Fix comment typo that managed to sneak in when I copy pasted some
comments & code from iostat.
2008-11-04 18:20:54 +00:00
Giorgos Keramidas
4fb6032f19 Repeat vmstat header after window.rows instead of a hardcoded 20.
Use ioctl() to get the window size in vmstat(8), and force a new
header to be prepended to the output every time the current window
size changes.  Change the number of lines before each header to the
current lines of the terminal when the terminal is resized, so that
the full terminal length can be used for output lines.

Inspired by:	svn change 175562 (same feature for iostat)
Reviewed by:	ru (who fixed some of my bugs too)
MFC after:	1 week
2008-11-04 18:02:35 +00:00
John Baldwin
32c9e3bd01 Use kvm_getcptime(3) to fetch the global CPU time stats from a crashdump
since the 'cp_time' symbol doesn't exist in recent kernels.  This fixes
iostat and vmstat on crash dumps.

MFC after:	1 week
2008-08-19 21:33:09 +00:00
Ruslan Ermilov
e83412c9ca Add forgotten -H, -h, and -P to usage(). 2008-04-10 09:56:57 +00:00
Peter Wemm
248e52ada8 Fix some boolean logic errors. && vs & and other sillyness. *blush*
This would prevent it from skipping non-present cpus in -P output.

Submitted by:  Pieter de Goeje <pieter@degoeje.nl>
2008-01-18 22:09:44 +00:00
Peter Wemm
4adb5266e8 Add the -H, -h and -P flags to vmstat. -P causes per-cpu output of
user/system/idle stats.  -h feeds the memory column through
humanize_number() to reduce the amount of column overflowing.  -H turns
this off.  -h is turned on by default if stdout is a tty.
2007-12-13 02:36:11 +00:00
Alan Cox
eaa29f1ce4 Add a counter for the total number of pages cached and support for
reporting the value of this counter in the program "vmstat".

Approved by:	re (rwatson)
2007-07-27 20:01:22 +00:00
Ruslan Ermilov
bad4d172b4 - Revert signedness type changes to "struct vmtotal"; by making
them unsigned I made the possible overflows hard to detect,
  and it only saved 1 bit which isn't principal, even less now
  that the underlying issue with the total of virtual memory has
  been fixed.  (For the record, it will overflow with >=2T of
  VM total, with 32-bit ints used to keep counters in pages.)

- While here, fix printing of other "struct vmtotal" members
  such as t_rq, t_dw, t_pw, and t_sw as they are also signed.

Reviewed by:	bde
MFC after:	3 days
2006-11-28 12:46:02 +00:00
Ruslan Ermilov
9623c97258 Oops, fix the format specifier to what was intended. 2006-11-23 11:47:34 +00:00
Ruslan Ermilov
5c88a11e6e - Fix types of "struct vmmeter" members so they are unsigned.
- Fix overflow bugs in sysctl(8), systat(1), and vmstat(8)
  when printing values of "struct vmmeter" in kilobytes as
  they don't necessarily fit into 32 bits.  (Fix sysctl(8)
  reporting of a total virtual memory; it's in pages too.)
2006-11-20 16:04:41 +00:00
Ruslan Ermilov
ddf3250846 - Add comma after REQUESTS field missed in previous commit.
- Widen some columns; make width of header columns less cryptic.
2006-10-21 10:16:26 +00:00
Gleb Smirnoff
d0c8105042 Print the number of allocation failures in UMA zones.
PR:	kern/102940
2006-10-21 09:08:35 +00:00
David E. O'Brien
bb6988aa71 Increase the field widths of flt (total number of page faults), fr (pages
freed), & cs (CPU context switch rate).  'vmstat 1' output is now lined up
for today's typical machines vs. a VAX.  [tested my modest 1.6ghz laptop]
2006-06-03 20:34:56 +00:00
Andre Oppermann
a21cbcb876 Obtain true uptime through clock_gettime(CLOCK_MONOTONIC, struct *timespec)
instead of subtracting 'bootime' from 'now'.

Sponsored by:	TCP/IP Optimization Fundraise 2005
2005-10-17 15:37:22 +00:00
Robert Watson
768528adf6 Use libmemstat(3)'s kvm support for malloc(9) rather than hand-extracting
this information from the core dump.
2005-08-06 13:56:21 +00:00
Robert Watson
c753d56b39 Teach vmstat's domemstat_zone() to use memstat_kvm_uma() when the kvm
descriptor is non-NULL, restoring vmstat -z support for core dumps and
kmem access.  These were broken with the introduction of UMA.
2005-08-01 21:33:17 +00:00
Robert Watson
b9dc1340d2 Minor syntax tweaks:
- Remove some extra blank lines.
- Remove comments that don't contribute to understanding.
- Remove additional blank lines in output added to maximize compatibility
  with older vmstat output, but that is actually somewhat gratuitous.

Submitted by:	bde
MFC with:	other vmstat libmemstat(3) changes
2005-07-28 09:45:08 +00:00
Robert Watson
8e73022dda Teach vmstat -m and vmstat -z to use libmemstat(3). Certain
statistics from -z are now a bit different due to changes in the
way statistics are now measured.  Reproduce with some amount of
accuracy the slightly obscure layouts adopted by the two kernel
sysctls.  In the future, we might want to normalize them.

GC dosysctl(), which is now no longer used.

MFC after:	1 week
2005-07-25 22:35:10 +00:00
Robert Watson
147819cc85 Modify vmstat(8)'s domem() routine, which is responsible for extracting
malloc(9) statistics from kernel memory or a kernel coredump, to catch
up with recent changes to adopt per-CPU malloc(9) statistics.  The new
routines walk the per-CPU statistics pools and coalesce them for
presentation to the user.
2005-05-29 13:40:00 +00:00
Ruslan Ermilov
f682f10c76 Sync program's usage() with manpage's SYNOPSIS. 2005-05-21 09:55:10 +00:00
Paul Saab
043da612df Fix an overflow when calculating the number of kilobytes from the
number of pages.

Obtained from:	Yahoo!
2005-02-21 14:35:00 +00:00
Brian Feldman
22ab6be3d4 Make vmstat -m work with -M/-N again. Note that making vmstat -z work
is much harder, and -m is grossly using unexported interfaces (that is,
the array of malloc zones/sizes does not have an exported type).
2004-06-30 01:34:26 +00:00
Tim J. Robbins
6d4ba9cd4e Print fork statistics with %u as they are unsigned quantities.
PR:		65889
Submitted by:	Ken Stailey
2004-04-23 13:10:29 +00:00
Ruslan Ermilov
ab82970ed2 Properly document the -M and -N options. 2004-03-26 11:01:45 +00:00
Bruce Evans
c044589554 Fixed a misspelling of 0 as NULL.
Fixed a nearby bug.  The "play it safe" code in dosysctl() was unsafe
because it overran the buffer by 1 if sysctl() filled all of the buffer.

Fixed a nearby style bug in output.  Not just 1, but 2 extra newlines
were printed at the end by "vmstat -m" and "vmstat -z".  Don't print
any newlines explicitly.  This depends on 2 of the many formatting
bugs in the corresponding sysctls.  First, the sysctls return an extra
newline at the end of the strings.  This also messes up output from
sysctl(8).  Second, the sysctls return an extra newline at the beginning
of the strings.  This is good for separating the 2 tables output by
"vmstat -mz" and for starting the header on a new line in plain sysctl
output, but gives a bogus extra newline at the beginning for "vm -[m | z]"
and "sysctl -n [kern.malloc | vm.zone]".

Fixed some nearby style bugs in the source code:
- the same line that misspelled 0 as NULL also spelled NULL as 0.
- the size was doubled twice in the realloc loop.
- the "play it safe" comment was misleading.  Terminating the buffer
  is bogus because dosysctl() is only meant to work with sysctls that
  return strings and the terminator is part of a string.  However, the
  kern.malloc sysctl has more than style bugs.  It also doesn't return
  a string.  Termination is needed to work around this bug.
2004-03-11 11:30:57 +00:00
David Malone
67206e04f9 Remove an unused variable.
Add some missing constness.
2003-12-25 01:32:55 +00:00
Bruce Evans
06849fd97a Fixed misplacement of __FBSDID(). Backed out editing of vendor id lines.
Just wrap them in #if 0...#endif.
2003-12-11 10:58:29 +00:00
Dag-Erling Smørgrav
56f9b09632 Revert part of revision 1.74 after bde reminded me of a detail I'd
forgotten about how sysctl works.  This removes a potential (though
not very likely) race that 1.74 introduced.
2003-12-11 07:46:08 +00:00
Dag-Erling Smørgrav
1b62c2410d Whitespace cleanup. 2003-12-10 22:19:19 +00:00
Dag-Erling Smørgrav
dc4ed4fe11 Remove debugging printf 2003-12-10 22:14:32 +00:00
Dag-Erling Smørgrav
7d268144f1 Fix a couple of issues in the interrupt code:
- Replace overly-complicated (and buggy) -a logic with a much simpler
   version: -a causes all interrupts to be displayed, otherwise only
   those that have occurred are displayed.  This removes the need for
   any MD code.

 - Instead of just making sure intrcnt is large enough, figure out the
   exact size it needs to be.  We derive nintr from this number, and we
   don't want to risk printing garbage.  Note that on sparc64, we end up
   printing garbage anyway because the names of non-existent interrupts
   are left uninitialized by the kernel.

Tested on:	alpha, i386, sparc64
2003-12-10 22:11:51 +00:00
Dag-Erling Smørgrav
3c455bcacc s/u_(int|long)\>/unsigned \1/ 2003-12-10 21:43:10 +00:00
David Malone
e38926c160 Fix a couple of warnings (const a function parameter and change some
ints to size_ts to better match the types of variables they are used
with).

Glanced at by:	des
2003-12-10 19:24:11 +00:00
Marcel Moolenaar
7e00080caa Unbreak vmstat -i on ia64:
o  nintr and inamlen must by of type size_t, not int,
o  Remove now unnecessary casts,
o  Handle the aflag differently, because the intr. names have a
   fixed width and almost always have trailing spaces.
2003-12-08 07:57:57 +00:00
Dag-Erling Smørgrav
c51633823d Finish the transition from libkvm to sysctl that I started a while ago.
The use of libkvm for post-mortem analysis is still supported (though it
could use more testing).  We can now remove vmstat's setgid bit.

While I'm here, hack the interrupt listing code to not display interrupts
that haven't occurred unless the -a option was given on the command line,
and document this change.
2003-12-07 17:46:14 +00:00
John-Mark Gurney
25d136776e make minimum column size for interrupt name be the heading (depends upon
Total being shorter than interrupt)..

Pointed out by:	bde
2003-11-09 20:39:56 +00:00
John-Mark Gurney
2eb7cf1a2b fix an overflow bug when scanning for length of names that I introduced in
the last commit...

include some minor style changes and fixes that bde sent me

Submitted by:	bde
2003-11-08 07:24:03 +00:00
John-Mark Gurney
a86e06a60c take mux's fix to the next level, scan the names and make the field larger
as necessary... on sparcs where:
atapci0 vec1996
is a bit too long
2003-10-28 05:02:03 +00:00
Tim J. Robbins
3c93cfb510 Reimplement the vmstat -f option (display fork statistics).
PR:		50924
Submitted by:	Ken Stailey (with a few changes)
2003-04-14 07:21:03 +00:00
Maxime Henrion
74d66b34d7 Slightly rework formatting in vmstat -i so that it doesn't
break with "fooX irqY" lines that are a bit bigger.
2003-04-10 00:44:18 +00:00
Poul-Henning Kamp
7194d335cf Run a revision of the devstat interface:
Kernel:

Change statistics to use the *uptime() timescale (ie: relative to
boottime) rather than the UTC aligned timescale.  This makes the
device statistics code oblivious to clock steps.

Change timestamps to bintime format, they are cheaper.

Remove the "busy_count", and replace it with two counter fields:
"start_count" and "end_count", which are updated in the down and
up paths respectively.  This removes the locking constraint on
devstat.

Add a timestamp argument to devstat_start_transaction(), this will
normally be a timestamp set by the *_bio() function in bp->bio_t0.
Use this field to calculate duration of I/O operations.

Add two timestamp arguments to devstat_end_transaction(), one is
the current time, a NULL pointer means "take timestamp yourself",
the other is the timestamp of when this transaction started (see
above).

Change calculation of busy_time to operate on "the salami principle":
Only when we are idle, which we can determine by the start+end
counts being identical, do we update the "busy_from" field in the
down path.  In the up path we accumulate the timeslice in busy_time
and update busy_from.

Change the byte_* and num_* fields into two arrays: bytes[] and
operations[].

Userland:

Change the misleading "busy_time" name to be called "snap_time" and
make the time long double since that is what most users need anyway,
fill it using clock_gettime(CLOCK_MONOTONIC) to put it on the same
timescale as the kernel fields.

Change devstat_compute_etime() to operate on struct bintime.

Remove the version 2 legacy interface: the change to bintime makes
compatibility far too expensive.

Fix a bug in systat's "vm" page where boot relative busy times would
be bogus.

Bump __FreeBSD_version to 500107

Review & Collaboration by:	ken
2003-03-15 21:59:06 +00:00
Poul-Henning Kamp
571d13be84 Update to use current rev of devstat API. 2003-03-15 21:25:44 +00:00
Poul-Henning Kamp
190ff54ebb Add #include <sys/resource.h>
My apologies for missing these #includes, I must have confused the
dependencies with a wrong timestamp or something.
2003-02-16 16:01:52 +00:00
Poul-Henning Kamp
f341ca9891 Remove #include <sys/dkstat.h> 2003-02-16 14:13:23 +00:00
Matthew Dillon
f7550ecf3f Make 'sysctl vm.vmtotal' work properly using updated patch from Hiten.
(the patch in the PR was stale).

PR:             kern/5689
Submitted by:   Hiten Pandya <hiten@unixdaemons.com>
2003-01-11 07:29:47 +00:00
Thomas Moestl
95ba428ccb The boottime variable in sys/kern/kern_tc.c is a struct timeval, not a
time_t, so do not use the latter as type when retrieving the variable
via libkvm. This should fix vmstat on sparc64.
2002-08-09 15:47:43 +00:00
Jeff Roberson
6f2671750e Remove malloc_type's ks_limit.
Updated the kmemzones logic such that the ks_size bitmap can be used as an
index into it to report the size of the zone used.

Create the kern.malloc sysctl which replaces the kvm mechanism to report
similar data.  This will provide an easy place for statistics aggregation if
malloc_type statistics become per cpu data.

Add some code ifdef'd under MALLOC_PROFILING to facilitate a tool for sizing
the malloc buckets.
2002-04-15 04:05:53 +00:00
Jeroen Ruigrok van der Werven
e2cf520799 Fix a missed conversion of lld to llu for the uint64_t ks_calls and cast
to unsigned long long.

Don't be too overzealous with the printing of ks_calls in the total
statistics, cut back from 20 to 13 positions to print (which should last
a couple of years easily (20 digits is enough for 3168 years of calls at a
measly billion (10^9) calls per second.)).

Submitted by:	bde
2002-04-13 10:32:36 +00:00
Jeroen Ruigrok van der Werven
9e74ac9de4 Cast totreq to unsigned long long, as needed on 64-bit machines.
Pointed out by:	jeff
2002-04-08 21:06:14 +00:00